Two-step uncertainty analysis of Watts Bar Nuclear 1 cycle 1 with SCALE/PARCS1

The ability to quantify uncertainties in modeling and simulation of nuclear systems is of great importance. One source of uncertainty stems from the uncertainty in nuclear cross sectional data. A common approach to reactor core analysis is the two-step method in which homogenized few-group lattice cross sections are generated by lattice physics calculations and used in a full core simulator. Two common codes used in the two-step method are the SCALE code package and the PARCS (Purdue Advanced Reactor Core Simulator). It is important to understand how uncertainty in cross sectional data affects the core calculation results. Cross section uncertainties in SCALE are found in the covariance libraries. Using the Sampler stochastic uncertainty analysis super-sequence, covariance data are used to generate perturbation factors and calculate perturbed self-shielded cross sections for lattice and depletion calculations. These perturbed cross sections are used by the Purdue Advanced Reactor Core Simulator to perform full core calculations. Using a large number of perturbed cross sections, we can begin to quantify how the uncertainty in cross sectional data affects the core calculation end result. This paper will discuss the results of this uncertainty analysis methodology applied to the Virtual Environment for Reactor Applications benchmark progression problems. (authors)